DESCRIPTION (adapted from the application) We propose a collaboration of imaging scientists and experts in B-cell biology to develop new imaging technology and agents to assess B-cell mass and function. Advances in in vivo imaging methods have not yet made an impact on study of the endocrine pancreas. Our premise is that this inherently chemical organ needs to be imaged with chemical resolution as well as structural resolution in order to advance research in B-cell biology and the pathophysiology of diabetes. Complementary nuclear medicine (PET) and magnetic resonance (MRS) methods will be applied to an in vitro artificial organ system and to small animal in vivo imaging studies of healthy and diabetic animals. New imaging tools will be developed and validated to answer some fundamental questions about the biology of normal and injured B-cells. In SA1 we will use principally MRS to determine the amount of viable cell mass in our in vitro organ and compare this measure with oxygen metabolism and insulin release. We anticipate that the 31P peak ratios will be a robust indicator of B-cell viable mass that is more sensitive than the insulin release assay and will permit direct comparisons between imaging agents for B cells and islets. In SA2 we will also use the artificial organ model to evaluate a series of compounds for their potential as imaging agents to quantitative B-cell mass in vivo. Because only a small fraction of the pancreas is B-cells, a successful imaging agent will require a high level of selectivity over acinar cells as well as cells of neighboring organs. Selectivity could derive from chemical specificity of a radiotracer and/or from a unique time course of the tracer in B-cells, as analyzed by mathematical models. Because there is no existing literature on radiopharmaceuticals with the desired specificity for the endocrine pancreas, we have searched broadly to find good candidate imaging agents for radiolabeling. We have identified selective B-cell toxins, dyes for in vitro staining of islets, unique B-cell surface receptors, and molecules that play a role in the cell's specific biochemical function related to insulin production and release. Good candidate molecules will be labeled with PET nuclides and used in SA3 to image the natural history of B-cell mass in animal models of both spontaneous and toxin-induced diabetes. Imaging using the small animal PET will be compared with functional measures, including insulin release and other assays developed in SA1. The application of non-invasive islet imaging will allow investigators to answer questions about the etiology and natural history of type I diabetes, to improve ways to predict who will develop clinical disease, and to various drugs in preventing progressive beta cell destruction.